Migration, migration, migration

You may have guessed it. This article is about football … kidding it is all about migration. Always difficult, always loaded with a lot of misunderstandings, high costs and so on and so forth. Customers don’t want to migrate as they don’t see a benefit. Costly with little to no new functionality which a customer either don’t need or want to use.

But why should you migrate anyway?

  • Keeps you up-to-date with fixed issues
  • Newest security updates from 3rd. party components to secure your data and therefore your business
  • If you miss out on minor versions you’ll pile up a growing mountain of “challenges”
    • Higher complexity as list of new functionalities and fixes grows
    • You get hung up on our overall progress, not just Intershop Commerce Management, like (partly work in progress)
      • Virtualization
        • Cloud
        • Docker
        • Ansible
      • Continuous Integration/Continuous Development (Intershop 7.4.6)
        • Gradle
      • Monitoring
        • Elastic stack
      • Microservice (Intershop Commerce Management 7.8)
        • JAX-RS
      • Service Registration & Service Discovery (Intershop Commerce Management 7.7)
        • Netflix Eureka
      • Just to name a view things in the pipeline …

OK, but what about costs and effort for a migration?

It’s been more than two years ago where a handful of guys from RnD had a meeting in Ilmenau (Intershop Office). This meeting was more, less the kick-off for our future architecture with the goal of better, easier migration. As far as I remember there we formed the goal of our migration efforts. And the goal was to get migration effort under 50 days. In the mean time a lot has changed since then including our direction with SaaS and B2X. Guess what it is even more important now to have a minimal effort for migrations. As a SaaS provider we can’t afford costly migrations and long running migration projects. It must be possible to swap the platform underneath without customers taking notice of it.
How far did we got with our 50 days max effort for a minor version migration? I personally can tell only by hearsay since ICM 7.5 the effort is in this region. With 7.6 and the introduction of the inSPIRED (responsive) Storefront this effort might have been raised significantly if customers decided to migrate to inSPIRED as well. Roughly 5 weeks itself for close to product migration from PrimeTech to inSPIRED is also not too bad. Not bad for this huge benefit anyway! But in general it isn’t a necessary thing to do so.
The good news though from ICM 7.7 is that a bigger customer managed to do a migration from 7.6 within 10 days. Not too bad either!

If you are convinced now that migration is a good thing and you don’t know how to start click here to find an implementation partner of your choice or contact us here.

Migration, migration, migration

One thought on “Migration, migration, migration

  • 2017-11-15 at 9:46 am
    Permalink

    Urgent call to ICM 7.5 customers:

    ICM 7.5 is the last product in support which uses an JDK 7. We discovered recently that JDK 7 doesn’t cope with newer TLS certificates. This means that if a 3rd. party interface is used from the Application Server which uses HTTPS with a newer certificate JDK 7 can’t handle those connections anymore.
    Please consider an update to at least ICM 7.6!

Comments are closed.